учебный год 2023 / (Encyclopedia of Law and Economics 5) Boudewijn Bouckaert-Property Law and Economics -Edward Elgar Publishing (2010)

Security interests, creditors’ priorities, and bankruptcy 275

of scope and scale which create gains from aggregating the obligations of many participants, although the literature has yet to identify or theorize how those economies might operate.

Interestingly, one of the principle contexts in which debt has been collectively issued in markets over the last 30 years has been in connection with securitization transactions which specifically seek to avoid the inefficiencies of bankruptcy procedures (Kettering, 2008) Indeed the growth and importance of securitization furnishes some of the strongest empirical evidence against theories which claim that involuntary collective collection processes are likely efficient, even for collectively created debt obligations.

3.Individual vs. collective legal remedies

Bowers (1990) argues that it is likely to be excessively costly to involve creditors B, C. . .N in disputes between the debtor and creditor A. The default creditor’s remedy provided under state law in the United States and as an individual remedy in most legal systems is, accordingly, to entertain a legal proceeding on behalf of an individual creditor against the individual debtor. The proceeding is designed to permit the court to take the individual circumstances giving rise to each debt into account in determining whether to grant a remedy, and, if so, what remedy to grant. In the case of bond issues or mass torts, typical civil procedure is grounded on the assumption that the special circumstances under which the debt obligation arose are identical as among all of the claiming creditors. When collective obligations are undertaken under contract, on the other hand, and the terms of the contracts do not vary as between the claimants, collective proceedings do not differ from individualistic ones, in theory. The circumstances underlying the collective obligation are identical as between the debtor and all the creditors so the collection process can take them fully into account.

4.Collective proceedings for heterogeneous claims

The creditors’ remedies schemes of many developed nations, however, have created bankruptcy style procedures, collective creditors’ remedy processes under which individually formed claims are processed in a collective proceeding. If the individual circumstances result in the denial of a remedy to creditor B, more assets may remain for distribution to creditor C. C thus has an interest in the outcome of the collection dispute between B and the debtor so that an efficient collection regime would offer C the opportunity to intervene in the proceedings whenever C felt the expense of participating was less than the value of the private gains his participation might create. Typical bankruptcy proceedings, however, require all creditors to participate, on pain of loss of all their collection rights.

276 Property law and economics

It is intuitively plausible that an efficient law would provide a process involving only the debtor and the individual creditor to enforce individualized debt contracts. The mystery to be explained by law and economics of bankruptcy law is, then, to show how and when it is efficient for the law to provide a collective process involving the heterogeneous claims not only of creditor A, but also creditors B, C, D. . .N in enforcing A’s claim against the debtor. A process which gives consideration to the singular features of each credit transaction imposes procedural costs on creditors who were not parties to the transaction. A collective procedure which ignores those individualized circumstances threatens to reduce the values created by the process of seeking optimal matches between borrowers and lenders. The mystery of bankruptcy law is that it creates but cannot solve this dilemma. To make an efficient cost saving choice ex post is to deter efficient ex ante transactional matching. Attending to the efficiency of each match, on the other hand, makes the process excessively costly for other creditors.

5.What might justify collective bankruptcy processes?

Kanda and Levmore (1994) reason that the collective process can be justified if economies of scale exist in the conduct of collection efforts so that creditors would agree to join and pay a share of the reduced costs. It is possible that there are a limited number of cases in which the efficient matching which occurs at the time the debt obligation is initially created establishes a relationship which, despite the individualistic motivations for creating it, possesses features which nevertheless create economies of scale from joining its enforcement with the enforcement of other claims against the same debtor. Unfortunately, however, the literature has yet to describe or theoretically derive the features of such a case. Bankruptcy law itself governs nearly all debt contracts, and does not appear to attempt either to define cases which possess the requisite scale economies or to limit its application to just such cases.

6.The collective action problem model

The most famous law and economics analyst of bankruptcy systems made his name by arguing that when a single debtor has multiple creditors, the creditors may have a collective action problem. The individual incentives of each are to act in ways contrary to the best interests of all (Jackson, 1982). He therefore proposed that the creation of a mandatory collective remedy, which reflected the terms of an idealized multiple-creditor contract to cooperate, could be justifiable as a means of overcoming the collective action problem. Bankruptcy law could be explained if it actually incorporated the terms of that idealized ‘creditors’ bargain.’ The literature developed in three directions from Jackson’s basic insight.

Security interests, creditors’ priorities, and bankruptcy 277

(a)Does the law correspond with the model? First, there is a question whether the quest for efficiency (for example providing a legal solution to the ‘common pool’ problem creditors face under his analysis) could plausibly be regarded as driving the substance of the actual, existing bankruptcy doctrine. This first line of inquiry was addressed by Jackson himself with his collaborator Douglas Baird. They first attempted to apply the creditors’ bargain heuristic to predict the features that an efficient bankruptcy statute would contain. In a series of pathbreaking articles, however, they discovered that the manner in which American Bankruptcy Judges were applying the provisions of Federal Bankruptcy legislation differed from those predicted by their efficiency hypothesis (Jackson, 1984, 1985; Baird and Jackson, 1984, 1985). Jackson and Scott (1989) attempted to account for the previously observed inefficiencies as an insurance mechanism, but conceded that the mechanism was unlikely to prove workable.

(b)Is there really a serious collective action problem after all – is the model itself theoretically sound? Research in the second direction challenged the soundness of Jackson’s initial analysis. The ‘Creditors’ Bargain’ logic grew from the premise of an assumed ‘common pool’ problem in which, when the debtor neared insolvency, assets seized and sold by creditor A tended to directly harm creditor B because insufficient assets were left behind to satisfy all remaining claims. This premise is more than simply distributional. Jackson argued, for example, that creditor A would not take the costs to B, C, . . ., N into account in making his decision to collect. Thus A would not avoid taking actions which destroyed synergistic values to the assets in the debtor’s portfolio. A creditor owed $100 might take a valuable earring whose stone could be sold for $100, even when it was a member of a matched pair worth $300 when kept and sold together. Such losses were avoidable, under Jackson’s analysis, by forcing all the creditors to act collectively. Thus, he argued, the nonbankruptcy system had created a system of perverse incentives which, on the occasion of insolvency, were cured by switching over to the bankruptcy model.

Influenced by empirical evidence that the adoption of bankruptcy law had resulted in zero payouts to the supposedly cooperating creditors, Bowers (1990) argued that Jackson’s view was one-sided, because it looked only at the incentives facing the creditors and either ignored the existence of the debtors, or assumed implicitly that debtors were completely passive. Bowers argued that debtors had both the means at hand and the incentive to avoid the losses Jackson predicted would result from perverse creditor common pool incentives.

Among the means available to a debtor under the nonbankruptcy system was the power to optimally liquidate its assets and, indeed,

278 Property law and economics

subsequent empirical research on the behavior of financially distressed firms (LoPucki and Whitford, 1993b; Gilson, 1996) shows that they do, in fact, conduct substantial asset liquidations. Once the debtor’s incentives are brought back into the picture, the hypothesized common pool problem fades, leaving us once again without a compelling economic justification for bankruptcy law.

(c) Are there other, better solutions to the collective action problem? The third line of argument spawned by the Jackson creditors’ bargain thesis took the law and economics literature in a new direction. It argued that whatever ex post collective action problem the nonbankruptcy system might create can be avoided by the use ex ante of optimal credit contracts (Adler, 1993b). Picker (1992), for example, justified the invention of security devices such as mortgages by showing that the adroit use of security could eliminate common pool problems. Bowers (1991) argued that the default terms of the existing nonbankruptcy system of secured and unsecured credit already provided the terms of optimal credit contracts, which tended to induce efficient distributions of the distressed debtor’s least critical assets first to its most vulnerable creditors, thus minimizing the size of aggregate distress losses.

Finally, a number of scholars began to argue that the common pool or other perverse incentive problems which might occur in the nonbankruptcy system could be handled by contract (Adler, 1994c). Since corporate debtors probably dominate the economic impact of the American bankruptcy system, corporate finance approaches to the problem of understanding bankruptcy law have argued that borrowing firms are capable of creating new kinds of securities with attendant options and covenants, which can obviate any common pool problem. Since this literature tends also to address the issue of alteration of contractual priorities ex post, it will be discussed below in connection with this last issue.

D.Remedies Based on Assets vs. Remedies based on creditor type (herein of the law and economics of security interests)

1.The legal justification for security

Simultaneously with the bankruptcy debate discussed above, the law and economics literature was also engaged in a vigorous debate about the justification for granting contracted-for priority rights to secured creditors. The legal view of the justification, as exemplified in Kripke (1985) and Carlson (1994) was that security, by reducing the credit risk borne by the secured lender, tended to make loans available that creditors would not make on any other basis and so security interests could be justified on the

Security interests, creditors’ priorities, and bankruptcy 279

same grounds that justified the extension of credit itself. Recent empirical studies (Berger and Udell, 1995, Mann, 1997a), confirm that lenders seek security interests in order to control the levels of moral hazard they face in lending to debtors who have limited amounts of equity financing. In what is, perhaps, the most important article in the law and economics literature concerning priority and security issues, Jackson and Kronman (1979) first developed and used the creditors’ bargain theory, so influential in the bankruptcy literature, to provide an economic explanation for the development and use of security devices. They argued that an aspect of the creditors’ collective action problem was a perverse tendency for creditors to expend duplicate efforts thus over-monitoring the debtor. Security, they concluded, if issued to the least efficient monitors, relieved them of the impulse to monitor, curing the perverse incentive, reducing unnecessary monitoring costs because only the most efficient monitors would have any remaining incentive to do so.

2.The secured debt puzzle

In 1981, however, Alan Schwartz showed that given some typical theoretical economic assumptions (completely informed, risk neutral creditors, with homogeneous expectations of the probability of default), the grant of security to a secured creditor tended to do more than just reduce risk to that lender (thus reducing the incentive to conduct duplicative monitoring). In fact, the grant transferred risks onto unsecured creditors who, under these theoretical assumptions, would demand as much compensation for accepting the transfer as the secured party was likely to give as a discount on the interest premium for being granted the security. The corollary as applied to the Jackson and Kronman monitoring thesis was that to the extent secured creditors could safely reduce their monitoring efforts, the grant of security correlatively increased the need for unsecured creditors to monitor, so that the theory could not predict any savings in aggregate monitoring costs either. Following Scott (1977), Schwartz argued security was in theory simply a zero sum game. If the confection of security interests is costly, then, the mystery is, why would debtors ever grant them when they stand to gain nothing by it? This analysis, Schwartz pointed out, was simply an application of the famous Modigliani and Miller Theorem (1958) of the irrelevance of capital structure. The argument created what he named and what has since been known as ‘The Puzzle of Secured Debt’ (Schwartz, 1984). Just as firms obviously invest much energy in designing and adapting their capital structures, they also issue secured debt in the teeth of theories which predict they will not. Schwartz concluded, however, that none of the then-existing theoretical explanations for the employment of short-term security devices could be

280 Property law and economics

squared with the empirical evidence, the patterns of secured lending actually observed.

3.Possible efficient puzzle solutions

The Schwartz thesis, that granting security is costly to debtors and gains them nothing, is, of course, contradicted by the observation that much secured lending and borrowing actually occurs. Shupack (1989) has observed that the costliness question is relative – that is, the Schwartz argument loses some of its punch if it is more costly to contract for unsecured lending than to take a security interest. Nevertheless, any explanation for the grant of security must probably show that the issuing of security is likely to be efficient, so that some private gains to the borrower and secured lender must be hypothesized.

Theories which explain the private gain as coming from externalizing risks onto uncompensated unsecured creditors, of course, are normatively unattractive. Schwartz (1984) examined several more benign explanatory answers proposed to his original puzzle, including those of Levmore (1982) (proposing that secured parties receive priority in payment for the external benefits their monitoring of the debtor confers on other creditors) and White (1984) (arguing that creditors differ in their levels of risk aversion so that security is arguably an efficient means of reducing risk to the most risk averse). Schwartz dismissed such theories which, by explaining the existence of private gains to borrowers, however, generate predictions that all borrowing will be conducted on a secured basis, such that all of every borrower’s available assets will be encumbered before any unsecured borrowing occurs. All such theories, Schwartz argued, will be embarrassed by the fact that much unsecured lending takes place to borrowers with unencumbered assets. It thus seems likely that an eventual persuasive theory will have to show that security is both costly and beneficial, or else that unsecured lending achieves previously unknown gains, in order for it to explain the observed mixture of types of borrowings. A number of such theories have been proposed, which argue that the institution of security is likely to be efficient, including Adler (1993a), Bowers (1991), Buckley (1986, 1992), Kanda and Levmore (1994), Picker (1992), Shupack (1989), Scott (1986), Stulz and Johnson (1985) and Triantis (1992, 1994, 2000). Without some strong empirical confirmation of any of the competing theories, however, none has yet commanded a general level of acceptance in the law and economics community. Indeed, the current majority view is probably that the debate over the puzzle of secured transactions has been inconclusive (Scott, 1997).

Nevertheless, work on solving the puzzle continues. Schwartz himself, the creator of the puzzle, has proposed that the grant of security interests

Security interests, creditors’ priorities, and bankruptcy 281

in favor of early lenders can be explained as an efficient way of permitting debtors to bind themselves not to engage in future financing which might dilute the value of earlier loans, which benefits borrowers without injuring creditors (Schwartz, 1997a). A still more recent model generalizes Schwartz’s logic, and entertains the assumption that later creditors must incur costs to discover rights granted to earlier lenders. It finds that in such a context several features of secured lending tend to be explainable as reducing the deadweight loss of those due diligence costs. Their model also expands to explain features of other inter-creditor relations found in the law of creditors’ remedies such as the veil-piercing exception to corporate limited liability and the law of fraudulent conveyances (Ayotte and Bolton, 2007).

4.Inefficient solutions to the puzzle

The inconclusiveness of the search for ‘benign’ explanations for security devices, which explain the use of security as justified by the creation of efficient outcomes, has led to a recent spate of arguments that the institution of secured credit is not only unproven as an efficient practice, but, on the contrary, is positively exploitative and thus inefficient. Schwartz (1981) had initially considered that security devices were employed by lenders and borrowers as means of exploiting creditors, like tort claimants, or consumers who were unsophisticated about the impact that security might have on their claims and who would therefore not increase the risk premiums they charged for becoming unsecured creditors. The distributive thesis, Schwartz argued, could only be proven by showing that firms whose creditors were likely to be unsophisticated were more apt to grant security interests than were borrowers whose other creditors were less easy to exploit. Such behavior should even be evident and the fact that it is not led him to dismiss the exploitation hypothesis. Nevertheless, LoPucki (1994) and Bebchuck and Fried (1996, 1997) have proposed that the priority extended to secured lenders be partially or wholly abolished to prevent the externalization of risk onto unsophisticated unsecured lenders. Warren (1997) and Klee (1997) argue that even though some measure of priority might be due to secured lenders, the law of security interests should ‘carve out’ an arbitrary percentage of the value of the collateral to be distributed to unsecured creditors. Scott (1994) has also suggested that the incentive structures inherent in the private law-proposing organizations which produced the American uniform law on security devices may also create perverse legal doctrine. Adler (1994a) argues that the Scott hypothesis results from a one-sided analysis. Harris and Mooney (1994) and Carlson (1986) have urged that in view of the inconclusive nature of the economic debate, that the institution of secured credit can be justified by resort to the

282 Property law and economics

historical legal theories which have been accepted by courts and lawyers. None of the exploitative or political theories of secured lending have yet gained general acceptance among scholars as explanations for the institution of security. The puzzle, thus, still remains to be solved, or, to say the same thing in another way, none of the existing theories which attempt to explain the institution has yet been proven correct.

5.Asset-based security as an individual remedy

Schwartz (1989) attempted to change the focus of the argument he himself had created by arguing for the grant of first priority over all of the debtor’s assets in favor of the borrowing firm’s earliest-to-lend financier. Under his analysis, the grant of priority, which is the functional equivalent of the grant of security in all the borrower’s assets, can be an efficient way for firms with good projects to signal that they differ from firms with poor prospects. Since the first-in-time priority scheme he proposed resembles the priorities created under the existing law of secured credit, his argument reduces to a plea that the priority system cut loose from the asset-based nature it carries under existing security device law and that a creditorbased priority scheme be substituted therefore. His argument for this change is based, in part, on his assessment that the process of tying public notice to particular assets in the current regime is unduly costly. Bowers (1995) discusses some theoretical reasons why filing systems might impose excessive costs. Kanda and Levmore (1994), on the other hand, argue that notice is especially important only in asset-based priority schemes and is thus not so important if priorities are based on creditor characteristics as Schwartz proposes. Schwartz does not address the inevitable aspect of his proposal, however, that it must necessarily trigger some sort of collective collection proceeding in almost every case of nonpayment. Any second- to-lend creditor whose contract entitles him to be first to collect (as for example when a short-term trade creditor is seeking to collect against a debtor who has financed himself with a relatively long-term loan from the first-to-lend financer) must involve the financier as well as the debtor in any claim that his debt ought to be satisfied out of any of the debtor’s assets, in all of which, under Schwartz’s proposal, the financier has a priority interest. Although Schwartz offered his proposal on a conceptual basis only and so cannot be faulted for not having worked through the multitude of legal details which its adoption would inevitably necessitate, it is difficult to imagine that the debtor could even voluntarily pay the second-to-lend creditor with assets in which the first-to-lend had a superior interest. This look into Schwartz’s proposal, on the other hand, does generate an explanation for the character of current security device law which is asset-based and not creditor-based. In an asset-based system, in which all the debtor’s

Security interests, creditors’ priorities, and bankruptcy 283

assets are encumbered, every unsecured creditor must deal with the prior secured lender in order to realize payment of his claim out of any of the debtor’s assets. In that kind of case, an asset-based priority system can be seen as requiring a collective determination of the relative rights of at least two creditors every time only one wishes to collect. Unlike a creditor-type based system, however, an asset-based system permits debtors to retain some unencumbered assets which creditors can resort to without having to trigger a collective proceeding. The choice of asset vs. claimant-based priority rules, then, can be seen as another aspect of the choice between adopting individualized vs. collective creditors’ remedies. Particularly when priority rules are claimantrather than asset-based and claimants exist in large classes (such as, for example, when there are many shareholders and many unsecured creditors who share strata of priority), for any individual member of any class to collect, a legal proceeding almost necessarily must involve all the members of the claimant’s own class, as well as all the members of any competing class in the process.

E.The economic analysis of creditor priorities

1.Nonbankruptcy first-in-time priority

Under nonbankruptcy law, once a creditor completes the involuntary collection process by having one of the debtor’s assets seized and sold, title to that asset passed to the buyer at the judicial sale, and the asset was no longer available to satisfy claims by other competing creditors. The asset was no longer part of the debtor’s property. This property theory-based system resulted in a ‘race’ system of priorities. The first creditor to seize the asset obtained priority over later seizing creditors. The nonbankruptcy system of individual remedies is, principally, one of temporal priority. Early lending secured creditors and early seizing creditors prevail over later ones. There are, however, exceptions. One of the chief deviations is in favor of later lending secured parties whose loans financed the acquisition of the collateral. They are granted a ‘purchase-money’ priority over earlier lenders claiming security interests in such after-acquired collateral. Certain statutory lien claimants also prevail over earlier perfecting secured lenders. In admiralty, there are many later-in-time but first-in-priority claims to interests in vessels. Lawyers probably deem the first-in-time priority as the general rule and the last-in-time priority cases as exceptional. Probably for that reason, the law and economics analysis has begun by attempting to understand the general rule first. Very little progress has been made in explaining the last-in-time priority cases. Mann (1996) is one analysis. Before discussing what there is on that score in the literature, therefore, we will first address the temporal priority scheme in general.

284 Property law and economics

(a) Capricious factors influencing racing outcomes The outcome of the race among unsecured creditors can be influenced by a variety of arbitrary factors. Among the most capricious is the variance among courts in timelags for obtaining judicial relief. The race is normally won by successfully completing a lawsuit, after which a judgment can be entered. The right to seize and sell the debtor’s assets is usually assertable only after judgment has been obtained. Ceteris paribus, then, creditors suing in jurisdictions which have a year or more delay between the time a suit is commenced and the time at which it will be called for trial, will be disadvantaged in the race as compared with competing creditors who are capable of maintaining their actions in venues with shorter trial calendars. Perhaps to eliminate the capricious effects of these arbitrary factors, the common law developed a set of devices under which a creditor can, at the time of commencing judicial proceedings, reserve an early place in the order of finish, conditional only on completing the judicial proceedings.

(b) Finishing-place reserving devices and investments in collection The race system’s prejudgment finishing-place reserving devices such as writs of attachment or sequestration, or notices of lis pendens, if freely available, would make the race among unsecured creditors more closely resemble the order in which their causes of action arose and thus susceptible to more reliable planning at the time credit is initially extended. The creditor making the first loan to become due would be more likely to become the first-in-line. In the last 40 years, however, the use of prejudgment writs has been restricted in the United States on constitutional grounds in a series of important US Supreme Court cases: Snaidach v. Family Finance Corp. 395 US 337 (1967); Fuentes v. Shevin 407 US 67 (1972); North Georgia Finishing, Inc. v. Di-Chem Inc. 419 US 601 (1975). The court found the writs objectionable, however, only on grounds that they invaded constitutional interests of the debtor. Their impact on the priority as among creditors was not attacked and it is conceivable that constitutional, prejudgment priority-reserving devices could still be designed to meet that need. Indeed, however, the recording of public notice of a security interest under Article 9 of the Uniform Commercial Code is an equally effective way of reserving a priority position at the time credit is negotiated and thus may have rendered the prejudgment collection writs superfluous.

Much unsecured credit is extended on a demand basis, however, so that an initial lending unsecured creditor cannot easily assure itself a head-start in the race system if it lends for a fixed term. Later demand-basis lenders will always have a head-start. Typical loan agreements attempt to enhance the likelihood of a more nearly even starting time, however, by permitting lenders to accelerate the due date upon adverse information, for example